All of lore.kernel.org
 help / color / mirror / Atom feed
From: Stefan Priebe - Profihost AG <s.priebe@profihost.ag>
To: Mark Nelson <mark.nelson@inktank.com>
Cc: Sage Weil <sage@inktank.com>,
	Alexandre DERUMIER <aderumier@odiso.com>,
	Gregory Farnum <greg@inktank.com>,
	ceph-devel@vger.kernel.org
Subject: Re: OSD Hardware questions
Date: Fri, 29 Jun 2012 14:35:07 +0200	[thread overview]
Message-ID: <4FEDA0FB.9010401@profihost.ag> (raw)
In-Reply-To: <4FED9370.5030700@inktank.com>


Am 29.06.2012 13:37, schrieb Mark Nelson:
> On 6/28/12 4:25 PM, Stefan Priebe wrote:
>> Am 28.06.2012 17:33, schrieb Sage Weil:
>>> Have you tried adjusting 'osd op threads'? The default is 2, but bumping
>>> that to, say, 8, might give you better concurrency and throughput.
>> For me this doesn't change anything. I believe the ceph-osd processes
>> are the problem. I mean i've 8 cores x 3,6Ghz and 4 ceph-osd processses
>> use around 80%.
>>
>>> "Mark Nelson" <mark.nelson@inktank.com> wrote:
>>>> It would be interesting to see where all your CPU time is being spent.
>>>> What benchmark are you using to do the random writes?
>>>
>>> Definitely. Seeing perf/oprofile/whatever results for the osd under that
>>> workload would be very interesting! We need to get perf going in our
>>> testing environment...
>> I have it working. But even a call graph of 10s is around 120 000 lines
>> long ?!
>>
>> Stefan
>
> What tool did you use to do the profiling?

i used the perf tool bundled with the kernel source. I can send you the 
perf.data file if you want. But i don't see anything interesting. But 
maybe i've just used perf the wrong way.

Stefan

  reply	other threads:[~2012-06-29 12:35 UTC|newest]

Thread overview: 26+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-06-27 13:04 OSD Hardware questions Stefan Priebe - Profihost AG
2012-06-27 13:55 ` Mark Nelson
2012-06-27 14:55   ` Jim Schutt
2012-06-27 15:19     ` Stefan Priebe
2012-06-27 17:23       ` Jim Schutt
2012-06-27 17:54         ` Stefan Priebe
2012-06-27 18:38           ` Jim Schutt
2012-06-27 18:48             ` Stefan Priebe
2012-06-27 19:10               ` Jim Schutt
2012-06-27 19:14                 ` Jim Schutt
2012-06-27 15:53     ` Mark Nelson
2012-06-27 17:59       ` Jim Schutt
2012-06-27 15:13   ` Stefan Priebe
     [not found]     ` <CAPYLRzj916kW=KLy3dMTVPJRoNtPMP_Ejz+YAxRUJ5jZc+HeMg@mail.gmail.com>
2012-06-27 15:28       ` Stefan Priebe
2012-06-27 16:00         ` Mark Nelson
2012-06-28 13:21           ` Stefan Priebe - Profihost AG
2012-06-28 14:38             ` Mark Nelson
2012-06-28 15:18               ` Alexandre DERUMIER
2012-06-28 15:33                 ` Sage Weil
2012-06-28 15:45                   ` Alexandre DERUMIER
2012-06-28 15:48                     ` Jim Schutt
2012-06-28 21:25                   ` Stefan Priebe
2012-06-29 11:37                     ` Mark Nelson
2012-06-29 12:35                       ` Stefan Priebe - Profihost AG [this message]
2012-06-28 16:01                 ` Stefan Priebe
2012-06-28 16:00               ` Stefan Priebe

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=4FEDA0FB.9010401@profihost.ag \
    --to=s.priebe@profihost.ag \
    --cc=aderumier@odiso.com \
    --cc=ceph-devel@vger.kernel.org \
    --cc=greg@inktank.com \
    --cc=mark.nelson@inktank.com \
    --cc=sage@inktank.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.